Mechanisms that control transcription elongation and termination are important components of gene expression in all organisms. We developed the B. subtilis trp operon leader as a model system for RNA polymerase (RNAP) pausing and intrinsic termination. We identified two pause sites in the B. subtilis trp leader (U107 and U144), which participate in transcription attenuation and translation repression mechanisms, respectively, although this has only been established for the translation repression mechanism in vivo. NusA and NusG stimulate pausing at both sites. Although NusA was known to stimulate pausing in E. coli, NusG- stimulated pausing is opposite to the anti-pausing activity identified fo E. coli NusG. NusG (Spt5) is the only universally conserved transcription factor. Since we identified the only examples of NusG/Spt5-stimulated pausing in any organism, we are in a unique position to investigate this novel NusG function. Our results indicate that NusG makes sequence-specific contacts with the non-template DNA strand within the paused transcription bubble. We will identify specific contacts between NusG and nucleic acids in the paused complex. We will also test a structural model of NusG-stimulated pausing by site-directed mutagenesis. Furthermore, we will use genomic approaches to identify NusG-stimulated pause sites throughout the B. subtilis genome to determine the prevalence of this pausing mechanism, which may be conserved in all three domains of life. While RNAP pausing is assumed to function in several attenuation mechanisms, this has never been shown for any attenuation mechanism in vivo. Using our previous U144 pausing studies as a guide, we will generate pause-defective mutants that will allow us to determine if pausing at U107 participates in the attenuation mechanism of the B. subtilis trp operon in vivo. Canonical intrinsic terminators consist of an uninterrupted RNA hairpin followed by a stretch of U residues. It has been known for many years that NusA is capable of stimulating termination at intrinsic terminators ~10-20%. Of particular interest, we found that NusA from B. subtilis and E. coli can greatly increase the termination efficiency at weak non-canonical terminators containing hairpin mismatches and/or poor U tracts (up to 18-fold). Although termination is not strictly dependent on NusA, we refer to this mechanism as NusA-dependent termination to distinguish it from the slight stimulation that occurs at canonical terminators. As hundreds of non-canonical terminators have been predicted in a variety of bacterial species, it appears that the number of terminators is far higher than previously thought. We will examine NusA-dependent termination in both B. subtilis and E. coli to determine if this previously overlooked termination mechanism is conserved in bacteria. Finally, we found that NusA-dependent termination regulates transcriptional readthrough into the nusA coding sequence in vitro. We will further characterize this novel autoregulatory attenuation mechanism that appears to rely on NusA-dependent termination rather than overlapping RNA structures.